f 
232 F.. A. P. Barnard on the Zodiacal Light. 
It is hardly necessary to point out how entirely different are 
these results from the facts of observation. If, on the other 
and, we suppose the light to be a mere locus of brightness, 
moveable laterally with the observer, then in this latitude, it 
should never be invisible either in the morning or in the evening 
ky :—at least, it should never be invisible unless the breadth of 
be consistent with the supposition that its source is so near to US. 
If the arc-radius of our horizontal circle is only 17° 14’, then | 
the preeeding limb of the limiting circle will, at the close of twl- 
light, have passed the zenith by 46/; and it will have to advance 
but 17° 14’—46/=16° 28’, in order to pass off from the horizon- 
tal circle entirely. The total duration of the phenomenon, there- 
fore, wp to the vanishing of the last trace of light, should be less 
than the duration of the preceding twilight—which, in this lati- 
tude is in the early part of March, less than an hour and a half. 
If we consider also that, long before it should entirely set, It 
would become inconspicuous and probably hardly visible, we 
shall see that it ought sensibly to disappear very soon after first 
presenting itself in the evening sky, instead of seeming to pat 
take, as it does, in its descent, of the general motion of the 
vens. 
It may be said, however, that we ought not to rest our reason- 
ings upon an assumption of so inconsiderable a distance for the 
ring ; since its visibility at midnight, and with the sun depre 
90°, demonstrates that it must be more distant. I am willing t 
take the largest distance that can be adduced from any observa 
tions ; or even a Jarger distance than would be required to satisfy 
any. We have seen that, if the light were visible in both hor 
zons at midnight, with a vertical altitude of 45°, it would poe 
duct us toa distance of less than 5000 miles above the earth's 
surface. Suppose therefore that we adopt 9000 miles as the tf 
ng, or rather, that of its highest visible stratum. The 
nee, if ina few particulars more favorable to the hy- 
